Repository navigation
Releases: himbeles/mermaid-render
Release list
v0.10.0
- Add editable Visio export for Mermaid sequence diagrams: participants, actors, lifelines, activations, glued messages, notes, and interaction frames.
- Export flowchart subgraphs as native Visio containers with nested membership and internal connectors.
- Preserve rendered node and connector stroke widths, and improve text and geometry compatibility.
- Expand examples, regression coverage, and offline wheel smoke tests.
v0.9.1
PNG rendering now defaults to scale 2 in both the CLI and Python API, producing twice the SVG's pixel dimensions (192 dpi). Use --scale 1 or scale=1 for the previous resolution.
The README now leads with PyPI installation and everyday usage, followed by offline bundled-wheel installation and contributing.
Validation: 80 tests passed, including PNG default-resolution and explicit-scale checks.
PyPI provides the lightweight universal wheel and source distribution. GitHub Releases also include offline bundled wheels for Linux x86-64, Windows x86-64, macOS Apple Silicon, and macOS Intel.
v0.9.0
Mermaid flowcharts now export to VSDX with editable right-angle connectors by default. Connections retain Mermaid's attachment positions, including points along diamond edges, while each shape also keeps its four standard connection points.
- Add
--visio-connectors right-angle|straight|mermaidand the Pythonvisio_connectors=option. Choosemermaidto retain detailed Bézier curves. - Preserve expanded flowchart shapes, including manual input and stacked documents/processes, as native editable geometry.
- Use
redux-coloras the default Mermaid theme. Preserve connector stroke colors, arrow-label backgrounds, node colors, and multiline text. - Fix self-loops, subgraph title placement and layering, and unwanted text backgrounds inside shapes.
Validation: 78 tests passed; application and expanded-shape examples were visually reviewed in an independent VSDX viewer. Editing/rerouting in Microsoft Visio remains unverified.
PyPI provides the lightweight universal wheel and source distribution. The GitHub release also includes offline bundled wheels for Linux x86-64, Windows x86-64, macOS Apple Silicon, and macOS Intel.